What Ensembl genome version should I use for alignments? (e.g. toplevel.fa vs. primary_assembly.fa)
这是一个很细节也很实际的问题，到底用哪个版本？
参考：
toplevel
ensembl
First part options:
- dna_sm - Repeats soft-masked (converts repeat nucleotides to lowercase)
- dna_rm - Repeats masked (converts repeats to to N's)
- dna - No masking
Second part options:
-
.toplevel - Includes haplotype information (not sure how aligners deal with this)
-
.primary_assembly - Single reference base per position
大部分都推荐使用soft-mask版本的，也就是没有把repeat替换为N。
下载hg19基因组：http://hgdownload.cse.ucsc.edu/goldenPath/hg19/bigZips/
参考：
从genecode下载hg19注释文件：ftp://ftp.sanger.ac.uk/pub/gencode/Gencode_human/release_27/
UCSC也可以下载，不过只能从网页导出。http://genome.ucsc.edu/cgi-bin/hgTables
注：genecode貌似出了问题，https://www.gencodegenes.org/releases/26lift37.html，里面ebi的链接无法下载了。
参考：http://www.biotrainee.com/thread-2035-1-1.html
基因组不是越新越好的，看看最新的CNS，里面很少有用最新版本的基因组，为什么？因为注释没跟上，你做出来的东西可能和别人对不上。
亲测
用不同版本的基因组效果会怎么样？
我做了转录组的测试，用的hg19和GRCh38
结论如下：
1. reads比对到基因组上的情况大致相同，基本没有差别；
2. 用不同的注释文件，基因表达的结果差距非常大。同样都是用featureCounts
GRCh38的结果：
Assigned 306852Unassigned_Unmapped 0Unassigned_MappingQuality 0Unassigned_Chimera 0Unassigned_FragmentLength 0Unassigned_Duplicate 0Unassigned_MultiMapping 36280Unassigned_Secondary 0Unassigned_Nonjunction 0Unassigned_NoFeatures 56950Unassigned_Overlapping_Length 0Unassigned_Ambiguity 19771
//================================= Running ==================================\\|| |||| Load annotation file /home/lizhixin/databases/ensembl/release91/Homo_s ... |||| Features : 1199851 |||| Meta-features : 58302 |||| Chromosomes/contigs : 47 |||| |||| Process BAM file /home/lizhixin/project/scRNA-seq/reanalyze/first_five ... |||| Paired-end reads are included. |||| Assign fragments (read pairs) to features... |||| |||| WARNING: reads from the same pair were found not adjacent to each |||| other in the input (due to read sorting by location or |||| reporting of multi-mapping read pairs). |||| |||| Read re-ordering is performed. |||| |||| Total fragments : 419853 |||| Successfully assigned fragments : 306852 (73.1%) |||| Running time : 0.05 minutes ||
hg19的结果：
Assigned 586467Unassigned_Unmapped 0Unassigned_MappingQuality 0Unassigned_Chimera 0Unassigned_FragmentLength 0Unassigned_Duplicate 0Unassigned_MultiMapping 66997Unassigned_Secondary 0Unassigned_Nonjunction 0Unassigned_NoFeatures 133437Unassigned_Overlapping_Length 0Unassigned_Ambiguity 47278
//================================= Running ==================================\\|| |||| Load annotation file /home/lizhixin/databases/cellranger_ref/refdata-c ... |||| Features : 1130716 |||| Meta-features : 32738 |||| Chromosomes/contigs : 45 |||| |||| Process BAM file /home/lizhixin/project/scRNA-seq/reanalyze/first_five ... |||| Paired-end reads are included. |||| Assign fragments (read pairs) to features... |||| Total fragments : 834179 |||| Successfully assigned fragments : 586467 (70.3%) |||| Running time : 0.05 minutes ||
不同的注释文件千万不要乱用！！！